Notebook for Info8000 Assignment3

The data was stored in csv format and loaded into R using “read_csv” command.

library(readr)
GA_SNAPstores_2008_2016_GeoID <- read_csv("D:/Dropbox/course/INFO 8000/Assignment3/GA_SNAPstores_2008_2016_GeoID.csv")

Parsed with column specification:
cols(
  .default = col_character(),
  SNAP_ID = col_integer(),
  NAICS_SHOR = col_integer(),
  SIZE_CODE = col_integer(),
  LARGE_CODE = col_integer(),
  ZIP5 = col_integer(),
  ZIP6 = col_integer(),
  LONG = col_double(),
  LAT = col_double(),
  TRACTID = col_double(),
  PUMAID = col_integer(),
  CTYID = col_integer()
)
See spec(...) for full column specifications.

head(GA_SNAPstores_2008_2016_GeoID,n=5)

Then we use “ggplot2” package" to draw a bar chart to visualize the distributions of SNAP-authorized store types.

ggplot(GA_SNAPstores_2008_2016_GeoID,aes(x=GA_SNAPstores_2008_2016_GeoID$STTYPE))+
  xlab("Store types")+
  ggtitle("Distribution of SNAP-authorized store types")+
  geom_bar(stat = "count")+
  coord_flip()+
  theme_bw()

As we can see, most of the SNAP-authorized stores are convenience stores.

Then we make a word cloud of the store names using “tm”,“wordcloud”,“RColorBrewer” packages.

#load needed packages
library(tm)
library(wordcloud)
library(RColorBrewer)
#create corpus
storename_corpus = Corpus(VectorSource(GA_SNAPstores_2008_2016_GeoID$NAME))
#build a term-document matrix
tdm = TermDocumentMatrix(storename_corpus,control = list(removePunctuation = TRUE,
                                                         stopwords = c("supermarket","shop","food","foods","mart","grocery","store","stop","llc","market","inc", stopwords("english")),
                                                         removeNumbers = TRUE, toupper = TRUE))
# define tdm as matrix
m = as.matrix(tdm)
# get word counts in decreasing order
word_freqs = sort(rowSums(m), decreasing=TRUE) 
# create a data frame with words and their frequencies
dm = data.frame(word=names(word_freqs), freq=word_freqs)
#visualize
wordcloud(dm$word, dm$freq, max.words=200, random.order=FALSE, colors=brewer.pal(8, "Dark2"))

Finally we use “leaflet” packages to visualize the locations of each store and explore the general pattern of it.Click the circle, then you can see the pop up window.

#load "leaflet" package
library(leaflet)
#define popups
map_popup <- paste0("<strong>Store Name: </strong>", 
                    GA_SNAPstores_2008_2016_GeoID$NAME,
                    "<br/><strong>Store Type: </strong>",
                    GA_SNAPstores_2008_2016_GeoID$STTYPE)
#Calculate mean center to set the view center
meanX<-mean(GA_SNAPstores_2008_2016_GeoID$LONG)
meanY<-mean(GA_SNAPstores_2008_2016_GeoID$LAT)
#Map showing all SNAP authorized stores locations
leaflet() %>% 
  addTiles() %>% 
  setView(meanX, meanY,zoom=7) %>%
  addCircleMarkers(GA_SNAPstores_2008_2016_GeoID$LONG,
                   GA_SNAPstores_2008_2016_GeoID$LAT,
                   radius = 0.5,
                   color = "#03F",
                   popup=map_popup)

If the points are too messy, we can use cluster option.Click the cluster circle, the cluster will zomm in.

leaflet() %>% 
  addTiles() %>% 
  setView(meanX, meanY,zoom=7) %>%
  addCircleMarkers(GA_SNAPstores_2008_2016_GeoID$LONG,
                  GA_SNAPstores_2008_2016_GeoID$LAT,
                  radius = 0.5,
                  color = "#03F",
                  popup=map_popup,clusterOptions = markerClusterOptions())